Testing data for randomness / predictability

Reply #32

lakerben — Sun, 12 Feb 2017 18:13:29 GMT

I get the same results after a tequila sunrise.

Reply #31

osmannica2001 — Sat, 11 Feb 2017 04:29:14 GMT

After so many years of analyzing data, have you been applying your y-values to combinatoric algorithms to simplify even further your output?

If visually some of us humans can see amazing patterns in drawings, why wouldn't a computer be able to visualize patterns even better? - just by using the right tools and simplified languages IMHO.

Reply #30

watch out — Sun, 29 Jan 2017 22:44:59 GMT

Off topic

Rl, I'm having pretty good luck bringing lines down from 120-200 usually end up with 8-12 lines with a lot of fours and a few 5's with the RAC.

settings

4

12

13

15

Reply #29

RL-RANDOMLOGIC — Sun, 29 Jan 2017 19:20:23 GMT

Why do you continue posting here, are you on drugs? Your post are incoherent and meaningless.

RL

Reply #28

CARBOB — Sun, 29 Jan 2017 17:11:19 GMT

That is the truest statement you can use to pick any filter that gives you 50/50.

Reply #27

Sunglasses — Sun, 29 Jan 2017 16:44:15 GMT

You are putting people in a dense fog. People can help you to make the fog thicker.

Reply #26

RL-RANDOMLOGIC — Sun, 29 Jan 2017 15:36:36 GMT

Everything considered I think analyzing digit by digit using the the 2-step tool and playing gut instincts.

I will play around with a few predictors for the steps as they seem to be easiest. I will take a 50/50 over

a 1 in 30 any day.

RL

Reply #25

RL-RANDOMLOGIC — Sun, 29 Jan 2017 15:24:46 GMT

These are the 4 games since I posted the data. The data as I posted were ordered according to last hit.

Games out has been recalculated.

01/28/17 123 out 57 games +4

01/27/17 211 out 61 games +3

01/26/17 111 out 11 games +2

01/25/17 212 out 53 games +1

There are 40 permutations of 0-1-2-3 for the first 3 rogue digits, remember the first digit is limited to 0 to 2.

I am still not sure how to apply this information in a beneficial way except for limiting the selection to those... [ More ]

Reply #24

Sunglasses — Sat, 28 Jan 2017 21:06:56 GMT

031 F- 53 F.AVG? - 28.84906 [F. S. 30] - 16 SKIPS - 11 14 12 13 24 25 16 65 1 43 1 64 0 76 19 27 54 26

020 hit - 49 avg - 30.95918 [ 30] - 14 skips - 16 11 31 7 27 0 11 10 11 8 86 43 49 15 16 33 4 38

012 hit - 48 avg - 32.3125 [ 30] - 21 skips - 54 70 7 65 56 3 27 23 32 56 37 4 51 58 18 31 47 8

021 hit - 48 avg - 32.25 [ 30] - 21 skips - 10 21 20 42 20 31 18 44 14 99 2 15 54 144 26 19 24

111 hit - 48 avg - 31.02083 [ 30] - 19 skips - 10 79 42 44 7 38 10 37 99 15 3 65 53 81 9

Reply #23

dr san — Fri, 27 Jan 2017 18:55:54 GMT

hello,A way of converting the digital data to the biological language, in which the binary numbers become corresponding between pairs of DNA sequences. In this way, the number 1 is identified by the sequences A and C, while the 0 by T and G.

Question = can someone do this in excel? to get

Pairs of sequences and trings

Reply #22

lottoswe — Fri, 27 Jan 2017 18:37:37 GMT

RL,

try this:

take median of rogues skips then find rogues with last 1,2,3 or 4 (or more) skips over median. Check when they are drawn. Check those on median too, if they are drawn in next draw.

lottoswe

Reply #21

RL-RANDOMLOGIC — Fri, 27 Jan 2017 14:35:50 GMT

Here are some stats on the hit rates for the first 3 rogues. The overall hit rate shows every combination

of three digits. The number of skips of 30 or more runs around 40 to 50%. Many games one can remove the

last 10 without much worry but that still leaves 30 possible choices. You can see that the combo 031 has

showed the greatest number of times at 53 and there are only 3 games with less than 11 skips between hits.

The average is 1 in 29 games and the hits greater that a skip of 3... [ More ]

Reply #20

dr san — Thu, 26 Jan 2017 11:30:37 GMT

excel, please

0 0 0 0

0 0 0 1

0 0 1 0

0 0 1 1

0 1 0 0

0 1 0 1

0 1 1 0

0 1 1 1

1 0 0 0

1 0 0 1

1 0 1 0

1 0 1 1

1 1 0 0

1 1 0 1

1 1 1 0

1 1 1 1

Reply #19

dr san — Thu, 26 Jan 2017 10:36:22 GMT

Hello, for pick4 it was much smaller than a vtrac, it was 100% in 16 formations

You can make statistics of the 16 trainings in 20 or 50 or 100 raffles to see the frequencies, from the last to the next does not repeat the training

We can try the last 3 raffles of pick4 not to repeat the training

Reply #18

dr san — Thu, 26 Jan 2017 10:31:06 GMT

Hello, where do we have to pick 4, for pairs

0,2,4,6,8 = 0

1,3,5,7,9 = equal 1

We have 16 trainings in 100%

Example

4568

5985

0 0 0 0

0 0 0 1

0 0 1 0

0 0 1 1

0 1 0 0

0 1 0 1

0 1 1 0

0 1 1 1

1 0 0 0

1 0 0 1

1 0 1 0

1 0 1 1

1 1 0 0

1 1 0 1

1 1 1 0

1 1 1 1

Reply #17

dr san — Wed, 25 Jan 2017 11:05:31 GMT

Hello, who is the guru of the wheels of the forum? It is thus trying to create wheels with standard DNA sequencing, ie has triad that does not suit certain traides

* That is a 49/6 with larger trios is likely to come out together, thereby creating wheels

* With a number and trios, trios and two numbers etc, as the pattern of DNA, it is possible

Reply #16

RL-RANDOMLOGIC — Wed, 25 Jan 2017 08:53:06 GMT

I'm with you, he makes no sense. In a 5 bit binary string there are 32 possible ways to sequence the 0's and 1's.

00000=0 to 11111 = 31. I think he's trying to make the point that just like in a coin toss the next value could be

either heads or tails and the outcome has nothing to do with what came before it, Bravo. The lottery is not truly

random, it's sufficiently random which is something else. In lottery games random is limited to the order of numbers

being drawn. For instance, i... [ More ]

Reply #15

Sunglasses — Tue, 24 Jan 2017 22:50:25 GMT

https://www.lotterypost.com/member/165235

Hello cocaratcha lottoswi, buenos noches, see combination s della lottoswa! Bueno! Mosquitos on the rockas

Reply #14

lottoswe — Tue, 24 Jan 2017 22:07:41 GMT

What?!

You sound like dr san

Reply #13

Sunglasses — Tue, 24 Jan 2017 19:00:52 GMT

Let's do a game. We play 1 or 0. You bet on the right outcome and you get your bet times 1.1 as win.

Five drawings can be one of those:

00000-00001-00010-00011-00100-00101-00110-00111-01000-01001-01010-01011-01100-01101-01110-01111-10000-10001-10010-10011-10100-10101-10110-10111-11000-11001-11010-11011-11100-11101-11110-11111

If you set up to play another five drawings, it is one of those above. If 00000 was drawn, and you are so pissed off to see no 1 coming up, that doesn't mea... [ More ]

Reply #12

lottoswe — Tue, 24 Jan 2017 17:59:10 GMT

Hi,

as i understand branch predictor collecting history of 00,01,10,11 and have some counter then decide what to choose. In lottery ,in first position is ,let's say this sequence

5,5,4,9,8,2,5,6,1,1,3,2. If 0 = even and 1 = odd then sequence is 1 1 0 1 0 0 1 0 1 1 1 0. Last two are 10 , 0 is first digit in next . There is only 01 or 00 with first digit 0. History counter will decide between 01 and 00.

That's how i see this.

ARL types looks like prediction professor Allan Lichtman sy... [ More ]

Reply #11

RL-RANDOMLOGIC — Mon, 23 Jan 2017 20:16:09 GMT

lottoswe

I took some time to read the article on the branch predictor and it's not the type of predictor one might experiment

with for lottery predictions. I am sure it could be modified to make some sort of lottery prediction tool but there are

many others that would be better suited to do the job. IMHO the best suited for lottery prediction is the Association

rule learning types. The best prediction tool I ever wrote used many association rules that were hard coded which

were th... [ More ]

Reply #10

RL-RANDOMLOGIC — Mon, 23 Jan 2017 12:52:46 GMT

After a series of test I am thinking the method might work better as a filter. The number of correct

predictions needed to reach the average budget would remove any advantage as any one miss will

remove the top level prize.

RL

Reply #9

RL-RANDOMLOGIC — Mon, 23 Jan 2017 01:58:31 GMT

lottswe

I have played around with many prediction tools and IMHO none of them are useful for predicting lottery numbers.

Random by definition is not predictable and the problem is this. Most prediction tools can work out the the best

solution to play based on the games history but we should not expect random to follow the algorithms output. In

other words the best has nothing to with what will happen next. Yes, some predictors do hit now and then but

even broken clocks show the co... [ More ]

Reply #8

lottoswe — Sun, 22 Jan 2017 22:20:55 GMT

Hi RL,

i found something called branch predictor but i don't know to much about it. I try to understand text on wiki. There is two level adaptive predictor sound interesting. If you know something about it, can you explain to me in layman terms how it working?

thank you

Reply #7

Sunglasses — Sun, 22 Jan 2017 15:08:03 GMT

You can find c-compilers on Internet, get a free student version that counts for non professional use. I don't use C as I didn't learn it, but it seems to be less complicated than j. What is, where is section 5, which pdf, which link to where? Evaluate binary string for randomness can only be returning its probability. There you are flipping coins, but when the 1111111111111111111111111 happens, don't complain that it isn't in the big middle chunk. Of course you can change the base too, then you... [ More ]

Reply #6

RL-RANDOMLOGIC — Sun, 22 Jan 2017 12:21:29 GMT

JW

Can you send me a link in a PM?. I not sure how much more I can gain using this vs the stuff I wrote but anything

is worth a look. I think the reason I could not download the file was because I was working on a XP machine.

RL

Reply #5

jimjwright — Sun, 22 Jan 2017 09:39:08 GMT

I modified the makefile to work with Visual Studio 2015 Professional, I assume it will work with Community Edition also. I can share it if anyone needs it.

Visual Studio probably requires at a minimum Windows 7 or later.

As Garyo said once you install Visual Studio Community Edition, you also have to do the following to install C++

https://www.youtube.com/watch?v=U1MRUiJAOq0

I am not familiar with any of the other older C compilers.

As far as your overall goal in theory it is prob... [ More ]

Reply #4

garyo1954 — Sun, 22 Jan 2017 08:11:53 GMT

Went ahead and tried Visual Studio Community. Defaults to C#.

After it downloads, installs, does the initial setup, and the initial start up you have the option to install other modules. MO, more headache than any one person should go through.

Go to the Microsoft Download Center. They have Visual Studio 2012 Pro and 2012 Ultimate.

Download Center

Reply #3

RL-RANDOMLOGIC — Sun, 22 Jan 2017 05:25:59 GMT

JW

It's been awhile, I tried downloading the express version of studio 2005 but the link seemed to be dead.

I was hoping to stay away from the gcc but looks like that may be the only way. My borland turbo C++

is older than the hills and think it's too dated. I was sure it would do the job but after editing the makefile

it reported 25 errors. I also tried to compile each test one at a time but not to be. I am thinking of porting

the code to qb64 but need to study it a little more.... [ More ]

Reply #2

garyo1954 — Sun, 22 Jan 2017 04:43:46 GMT

Agree with jim

Reply #1

jimjwright — Sun, 22 Jan 2017 04:40:48 GMT

It looks like the makefile is referencing gcc which is the gnu c compiler so you can try the following link.

https://gcc.gnu.org/

In the past it was a pain to set up gcc on a windows computer but maybe its simpler now. I think you had to install Cygwin first but its been a long time since I installed gcc. I exclusively develop on Mac now for last 3 years. Go to you tube and search for installing gcc on windows if you want to go this route.

I was able to compile as is without any modific... [ More ]

Testing data for randomness / predictability

RL-RANDOMLOGIC — Sun, 22 Jan 2017 03:59:58 GMT

Hi all

In the past I have tested many secondary data sets to measure it's predictability using my own tools

but a few weeks ago I ran across a battery of tools that may work better than what I am using. Here

is a link to the NIST webpage http://csrc.nist.gov/groups/ST/toolkit/rng/stats_tests.html The National

Institute of Standards and Technology.

First I would like to know if anyone has a link to a free C compiler that will compile the tools available for

download. I have turb... [ More ]